You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mesos.apache.org by "Benjamin Mahler (JIRA)" <ji...@apache.org> on 2013/10/23 21:27:41 UTC
[jira] [Resolved] (MESOS-762) Revert the use of the soft limit and
memory threshold notifications.
[ https://issues.apache.org/jira/browse/MESOS-762?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Benjamin Mahler resolved MESOS-762.
-----------------------------------
Resolution: Fixed
Fix Version/s: 0.15.0
Commited:
* 38254e7 Revert "Added memory threshold notifications for inducing OOMs."
* 8367d94 Revert "Updated the balloon tests to test the memory threshold."
* be9c22f Revert "Added a missing include."
* b347190 Revert "Fixed ARG_MAX compilation issue on some POSIX systems."
> Revert the use of the soft limit and memory threshold notifications.
> --------------------------------------------------------------------
>
> Key: MESOS-762
> URL: https://issues.apache.org/jira/browse/MESOS-762
> Project: Mesos
> Issue Type: Bug
> Components: isolation
> Affects Versions: 0.14.1
> Reporter: Benjamin Mahler
> Assignee: Benjamin Mahler
> Priority: Critical
> Fix For: 0.15.0, 0.14.2
>
>
> The use of the cgroup soft limit and threshold notifications to induce our own OOMs does not behave the same as relying on the hard limit and oom notifications.
> We previously used the cgroup hard limit which ensures a reclaim of cache memory is performed when the hard limit is reached. The latest version 0.14.1 switched to using the cgroup memory soft limit, which does not perform a reclaim when reached. The result is that executors that perform substantial I/O will build up a large amount of cache memory, eventually reaching the soft limit and killed as a result.
> The fix is that we're changing back to using the hard limit, to get back to the previous OOM behavior in Mesos.
--
This message was sent by Atlassian JIRA
(v6.1#6144)